JP2016527612A - Ranking comments by search engines - Google Patents

Abstract

The present disclosure provides an exemplary method and apparatus for performing comment ranking by a search engine. One or more terms that describe the object are extracted from the comments about the object. One or more useful attributes included in the comment are acquired according to a term describing the object. A comment scoring index is determined according to the number of useful attributes. Comments are scored according to the indicators. Comments are ranked according to the score. The technology allows users to view useful comments quickly and efficiently, and helps them make informed decisions.

Description

(Cross-reference of related applications)
The present application is Chinese Patent Application No. 201301425222 entitled “Method and Apparatus for Implementing Comment Ranking by Search Engine” filed on June 19, 2013, and which is entitled “Method and Apparatus for Implementing Comment Ranking by Search Engine”. Claims foreign priority over No. 8, the entire contents of which are incorporated herein by reference.

  The present disclosure relates to the Internet field, and more particularly, to a method and apparatus for ranking comments by search engines.

  With the development of the Internet in the world, more people are accepting the Internet and relying on the Internet in their work and daily life. Activities that previously required face-to-face interaction are gradually being replaced by the Internet.

  The development of the Internet has created various applications, one of which is electronic commerce, that is, e-commerce.

  E-commerce uses computer technology, network technology, and telecommunications technology to achieve the digitization, digitization, and networking of all transaction processing. The term e-commerce typically refers to a new business model that conducts global commerce and trade activities in an open network environment such as the Internet. In e-commerce, as well as all kinds of business activities, trade activities, economic activities, and integrated service related activities, various commercial and trade activities such as online shopping by customers, online transactions between vendors, online payments, etc. It is based on browser / server applications that the seller does not need to meet directly.

  With the development of e-commerce, more customers or users are choosing to purchase various products and services online, such as clothes, digital products, home appliances, books, lotteries and the like. However, it is not easy for users to select the most satisfying items from a wide range of goods and services on the Internet, especially when users are unable to see physical entities or are not familiar with the merchant. The case is not easy.

  To help users make decisions, many websites provide information about goods and services. Among them, the most convenient and effective is a website that provides other users' comments regarding the product or service. Accordingly, the user can make a decision based on a large number of comments about goods or services from a large number of other users.

  A comment search engine based on user comments is an important solution to solve the problem. When the user clicks on a product or product attribute on the web page, the comment search engine searches the index file according to the keyword and returns the most relevant comment information.

  However, due to the large number of comments regarding the desired product or desired service, the user will not have enough time and energy to read through all the comments. Therefore, the issue is how the user can quickly and efficiently obtain useful comments that help make a decision.

  Furthermore, such a comment creation platform is freely used by the user. In other words, the content of the comment is not restricted. As a result, many spam-like content such as advertisements, content unrelated to the product, or information content that is hardly useful can be included in the ever-growing number of comments about the product. Therefore, how to help users effectively delete or ignore comments containing spam content also becomes an issue.

  The prior art has solved these problems by ranking comments according to time. Specifically, the newest comment is displayed at the top, and the early comment is displayed at the bottom. In this way, the most recent comments are read to the user with the highest priority, and early published comments are read with a lower priority when the user scrolls down or changes pages. .

  However, the user usually begins to feel frustrated with the operation of changing the list page during the browsing process. According to the previous results, the user often stops browsing when entering the third page browsing, and the more the number of pages viewed, the higher the percentage of users who stop browsing the page.

  Therefore, good content published early to help the user's decision may not be read by the user. Therefore, the conventional sorting method that considers only the order of the publication times does not help the user to quickly get to high-quality content, and helps to reduce the time required for the user to make a decision. Absent.

  In addition, users often do not want to view each comment one by one, but rather want to focus on comments about specific attributes or specific aspects of a product or service. Such a request can be realized by performing an emotion analysis of a comment.

  The emotion analysis of a comment refers to performing a structural analysis on the content of the comment so that the user acquires an explanation of the entire product expressed through his / her comment and an attribute of each characteristic.

  Therefore, the present technology efficiently provides high-quality and useful comment contents to the user, and reduces the number of pages changed by the user.

  This summary is provided to introduce a selection of concepts in a simplified form that are further described below in the Detailed Description. This summary is not intended to identify every important or essential feature of the claimed subject matter, nor is it intended to be used alone to help determine the scope of the claimed subject matter. For example, the term “technology” may be used in the above context and as permitted throughout this disclosure for one or more devices, one or more systems, one or more methods, and / or multiple computers. Can refer to readable instructions. This technology improves the diversity of results in management technology.

  The present disclosure provides a method and apparatus for performing comment ranking by a search engine.

  The present disclosure provides an exemplary method for ranking comments by a search engine. One or more terms that describe the object are extracted from the comments about the object. One or more useful attributes included in the comment are acquired according to a term describing the object. Depending on the number of useful attributes, one or more scoring indicators for scoring comments are determined. Comments are scored according to the scoring index. Comments are ranked according to the score.

  The present disclosure also provides an exemplary apparatus for ranking comments by a search engine. The extraction module extracts one or more terms that describe the object from comments about the object. The acquisition module acquires one or more useful attributes included in the comment according to a term describing the object. The determination module determines one or more scoring indicators for scoring comments according to the number of useful attributes. The scoring module scores comments according to the indicators. The ranking module ranks the comments according to the score.

  Compared to the prior art, the present technology allows users to read useful comment information quickly and effectively. Users can better understand products, services, and merchants, thus allowing for better shopping decisions. This technology not only reduces the number of pages that users change when viewing comments and increases the efficiency of online shopping, but also helps websites effectively manage comments about their products and services .

The drawings are used to provide examples of the disclosure and are therefore part of the disclosure. Illustrative examples and descriptions thereof are used to illustrate the present disclosure, but should not be used to limit the present disclosure. In the drawings, the same reference numbers refer to the same or similar components.
6 is a flowchart illustrating an exemplary method for implementing comment ranking. It is a flowchart which shows the exemplary detailed operation | movement in FIG. 2 is an exemplary display of a web page according to an exemplary embodiment of the present disclosure. FIG. 3 is a schematic diagram illustrating an example apparatus for performing comment ranking according to an example embodiment of the present disclosure.

  Regarding ranking of comments (or views), the prior art does not consider the following factors. The factors are the amount of useful product attributes in the comment (ie the amount of useful information in the comment), the readability of the sentence (too long or redundant sentences are inconvenient for the user to get useful information) ), Amount of useless attributes in comments (attributes related to logistics or advertisements, etc.), and product attribute fields (waiting time is a very important product attribute on mobile phones, but it seems to be meaningless in products such as lotions. In addition, if the product changes, its useful attributes will also change).

  For ranking comments, the technology considers the amount of useful information in comments based on sentiment analysis. The amount of useful information in comments based on sentiment analysis is the amount of information that evaluates useful attributes, that is, useful information in comments, by dividing comments into attribute granularity levels through sentiment analysis, obtaining evaluation of useful attributes in comments Obtained by calculating the amount of information.

  For example, in a comment about a specific lotion, the sentence “Bought for a classmate” is useless without any evaluation of the product. Conversely, another sentence “This lotion has a good moisturizing effect” provides an evaluation of useful attributes for the moisturizing effect of the product, and is therefore useful information for the user.

  In this way, information that actually affects the user's understanding of the object, that is, useful information in the comment, is given a greater weight in the comment ranking process. When the useful information is considered, the display order of the comments is rearranged so that the content including more useful information is displayed at a higher priority position. Therefore, the number of pages required for the user to check the comment can be reduced. This technique not only saves the user effort, but also allows websites to manage and use comments more effectively.

  In an exemplary implementation of the present technology, the computing device comprises one or more processors or central processing units (CPUs), input / output interfaces, network interfaces, and memory.

  Memory includes non-permanent memory, random access memory (RAM), and / or non-volatile memory, such as read only memory (ROM) or flash memory (flash RAM), as used in computer readable media. It is. The memory can be viewed as an example of a computer-readable medium.

  Computer-readable media includes permanent and non-permanent, removable and non-removable media that can achieve the purpose of information storage by any method or technique. The term information is referred to as computer-executable instructions, data structures, program modules, or any type of data. Examples of computer readable or computer storage media include phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read Dedicated memory (ROM), electrically erasable programmable read only memory (EEPROM), flash memory or any other memory technology, compact disc read only memory (CD-ROM), digital versatile disc (DVD) or any other Optical storage media, cassette tapes, floppy disks or any other magnetic storage device, or any other non-transmission media used to store information and accessed by computing devices, Is not limited to . As defined herein, computer-readable media does not include transitory media such as modulated data signals and carrier waves.

  In order to make the objectives, technical solutions, and benefits of the present disclosure clearer, a detailed description of the present disclosure is provided below along with exemplary embodiments and accompanying drawings.

  In the following description, the use of "exemplary embodiment", "said exemplary embodiment", "example", "said example", etc., may have particular characteristics, structures, properties, characteristics, elements, or limitations, It may be shown to be included in certain exemplary embodiments or examples. Nevertheless, each embodiment or example need not include such specific features, structures, characteristics, characteristics, elements, or limitations. Further, the phrase “in an exemplary embodiment” that can refer to the same embodiment is used repeatedly, but does not necessarily refer to the same embodiment.

  In the following description, some technical features well known to those skilled in the art are omitted for the sake of brevity.

  FIG. 1 is a flowchart illustrating an exemplary method 100 for implementing comment ranking in a search engine, according to an exemplary embodiment of the present disclosure.

  At reference numeral 110, one or more terms that describe the object are extracted from comments about the object. The term “object” refers to an object associated with a comment.

  For example, in an e-commerce application, a user may provide comments regarding a specific product or service. Therefore, the object in the present embodiment is a specific product or service. Sometimes users may provide comments about different merchants for reference to other users. In this case, the object is the trader itself. In other embodiments, at a particular rating website, the user may provide comments regarding various restaurants. In this case, the object of comment is various restaurants.

  In other embodiments, at a particular news website, the user may provide comments regarding a particular news article. In this case, the object is a news article.

  In other examples, a user may provide comments on the status of their friends, other users, or content posted by friends on social networks. In this case, the object is the status of his / her friend on the social network, the content posted by another user, or the friend.

  It should be noted that the foregoing examples of objects are not limiting and are merely exemplary. Other embodiments of the term “object” in accordance with the present disclosure will be readily found by those skilled in the art.

  Terms used by users in comments in this disclosure to describe objects such as goods, services, or merchants are cited as emotional terms.

  For example, first, word division processing is applied to a user's comment. The word division process divides a sentence into terms. There are various methods for word division, such as the maximum forward match method, the maximum backward match method, and the maximum forward and backward match method.

  Then, according to the term acquired from the word segmentation process, a search is performed in a list including emotion terms managed on the server side, which is also known as an “emotion term dictionary”, and an emotion term included in the comment is specified. Is possible.

  At reference numeral 120, one or more useful attributes included in the comment are acquired in accordance with a term describing the object.

  For example, the term “useful attribute” as used herein refers to an attribute that is important to most users with respect to objects such as merchandise. Useful attributes vary with different types of objects. For example, useful attributes differ depending on products belonging to different fields and categories. In the case of a product such as a mobile phone, useful attributes include standby time, screen, functionality, and the like. In the case of different products such as lotion, useful attributes include moisturizing effect, whitening effect and nutritional effect.

  According to exemplary embodiments of the present disclosure, the useful attributes of an object can be obtained using data mining technology that extracts them from a large number of existing comments about the object.

  According to other exemplary embodiments of the present disclosure, the useful attributes can be preset.

  The useful attribute is an aspect (or characteristic) of an object such as a product, a service, or a business.

  Emotional terms refer to thoughts and views expressed by the user in comments about the object. For example, with regard to the useful attribute “nutrient effect”, the emotional terms used in the user comments may be words such as “nutritive”, “dry”, “great”.

  For example, in the lotion category, useful attributes include the entire product, nutritional effects, stimulation, and the like. Accordingly, the correspondence between standard useful attributes, useful attributes, standard emotion terms, and emotion terms is shown in Table 1 as an example.

  For the relationship between the standard useful attribute and the useful attribute, and the relationship between the standard emotion term and the emotion term, reference may be made to FIG.

  For example, suppose a user comment on a particular lotion is “I like this lotion in general. It has a light aroma, nourishment, and gentleness”. “Quite like” in the comment is an emotional term, “Generally” is a useful attribute corresponding to the standard useful attribute “whole product”, “Nutrition” is an emotional term, The term of the corresponding useful attribute is not included in the comment, and the standard useful attribute corresponding to the emotion term is “nutrient effect”. Thus, emotional terms sometimes appear in comments without the corresponding useful attribute terms directly.

  In a general situation, the corresponding useful attributes can be obtained directly from emotional terms.

  In some situations, it is not possible to obtain the corresponding useful attribute directly from the emotional term. For example, when the user comment includes the expression “both amount and quantity are valid”, the emotion term is “valid” and it is impossible to determine the corresponding useful attribute therefrom. Here, a search for the corresponding useful attribute of the emotion term can be performed in the comment. One useful attribute “amount” is acquired, and the corresponding standard useful attribute “price” is acquired. Another useful attribute called “quantity” is obtained, and the corresponding standard useful attribute “quantity” is obtained.

  In other situations, useful attributes cannot be obtained from emotional terms and the corresponding useful attributes may not be found in the comment. In other words, there are no useful attributes that objectively correspond to such emotional terms. For example, assume that an emotional term “sharp” is found in a user comment regarding a lotion. Such emotional terms do not have corresponding useful attributes. In this case, emotional terms are ignored without further analysis.

  According to an exemplary embodiment of the present disclosure, a table that records a correspondence relationship between at least two standard useful attributes, useful attributes, standard emotion terms, and emotion terms is retrieved on the server side during the comment ranking process. Can be held in a database.

  At reference numeral 130, one or more scoring indicators for scoring comments are determined according to the number of useful attributes.

  FIG. 2 is a more detailed flowchart illustrating the process at reference numeral 130 in FIG. At reference numeral 210, the technique determines whether the number of useful attributes in the comment is one or more.

  When the number of useful attributes in the comment is 1 or more, the process of reference number 220 is performed. At reference numeral 220, the technique determines a scoring index that includes the number of useful attributes.

An exemplary formula for calculating the number of useful attributes included in a comment is as follows:
SD = γ (1)
In the formula, γ represents the number of useful attributes obtained from the process of reference numeral 110.

  According to an exemplary embodiment of the present disclosure, when the scoring indicator includes the number of useful attributes, the scoring indicator further includes at least one of a comment readability, a useless attribute in the comment, and a comment granting time. Items can be included.

  The readability of a comment refers to an index as to whether or not the comment is easy for the user to read. Normally, a sentence that is too long, a sentence that is too short, or a sentence that has too many phrases is considered a sentence that is difficult for the user to read. As a result, the level of readability is lowered. According to an exemplary embodiment of the present disclosure, an exemplary formula for calculating the readability of a comment is as follows:

  In the formula, the symbol NL represents the standard length of all clauses in the user comment, and the length represents the number of characters in a particular clause of the user comment. A phrase is part of a sentence and is delimited by punctuation marks. According to an exemplary embodiment of the present disclosure, the number of characters in one Chinese character is 1, the number of characters in an English alphabet is also 1, and the length of the term “my” is 2, for example. In the formula, α represents an appropriate average value for a sentence that is easy to read for ordinary people, and β represents an appropriate variance for a sentence that is easy for ordinary people to read. α and β are both constants. Formula (2) represents the readability of the sentence using the standard length of the user comment.

  The useless attribute of a comment refers to an attribute that lowers the score related to ranking of comments.

  According to exemplary embodiments of the present disclosure, useless attributes can be predetermined and applied to all categories, or can be set based on experience or actual usage.

  For example, useful attributes are used for analysis purposes and displayed to the user, while useless attributes are only used for ranking. The user may click on the useful attribute on the web page so that the comments are rearranged in an order according to the useful attribute.

  According to an exemplary embodiment of the present disclosure, when a comment includes a useful attribute, an exemplary formula for calculating a ranking factor for the useless attribute of the comment is as follows:

  In the formula, the symbol FS represents a rank factor of useless attributes, ε represents the number of useless attributes included in the comment, and δ and ε are constants. The constants δ and ε have constant values, and are determined through a plurality of attempts so that the score of the following formula (5) decreases as the value of ε increases. Useless attributes can also be manually defined according to different ranking requirements. For example, most of the attributes displayed on some instant messaging tools are advertising attributes that can be considered useless attributes. If the application does not want excessive logistics information, attributes such as logistics service speed or logistics service customer service are considered useless attributes. Unnecessary attributes are generic to all products, are used to calculate ranking scores, and are not used to display attribute tags to the user.

  When the number of useful attributes in the comment is 0, the process of reference number 230 is performed. At reference numeral 230, the scoring index is determined to include an index other than useful attributes. According to the exemplary embodiment of the present disclosure, the scoring index other than the index of the useful attribute may include at least one item among the readability of the comment, the rank factor of the useless attribute in the comment, and the comment granting time. .

  The useless attribute of a comment refers to an attribute that lowers the score related to ranking of comments. In accordance with an exemplary embodiment of the present disclosure, if a comment does not include a useful attribute, an exemplary formula for calculating a ranking factor for the useless attribute of the comment is as follows:

  In the formula, the symbol FSN represents a rank attribute factor of useless attributes, ε represents the number of useless attributes included in the comment, and δ and ε are constants. The constants δ and ε are constant values, and each is determined through a plurality of attempts so that the score of the following formula (6) decreases as the value of ε increases.

  At reference numeral 140, the comment is scored according to the scoring index. In scoring, scoring indicators such as the number of useful attributes, readability of sentences, and useless attributes of sentences can be considered in the scoring process.

  According to exemplary embodiments of the present disclosure, if the comment includes useful attributes and the scoring indicator includes a ranking factor for the number of useful attributes, the readability of the comments, and the useless attributes in the comments, the final ranking score is An exemplary formula to calculate is as follows:

  In the formula, the symbol SD represents the number of useful attributes, the symbol NL represents the standard length of all clauses in the user comment, and the symbol FS represents the rank factor of the useless attribute calculated according to formula (3). .

  According to another exemplary embodiment of the present disclosure, the comment does not include a useful attribute, and the scoring index includes one or more scoring indices other than the useful attribute. The scoring index other than the useful attribute may include the readability of the comment and the rank factor of the useless attribute in the comment. In this case, an exemplary formula for calculating the final ranking score is as follows:

  In the formula, the symbol “num” represents the number of clauses, θ is a constant, which is determined through multiple attempts so that the ranking results are generally as expected. The symbol score represents the ranking score. The symbol NL represents the standard length of all clauses in the comment, calculated according to formula (1), and the symbol FSN represents the rank factor of the useless attribute calculated according to formula (4). As shown in formula (4), the greater the number of useless attributes, the more negatively affected the comment, or the lower the score corresponding to the comment.

  According to an exemplary embodiment of the present disclosure, commenting time is considered a factor in the process of calculating the final ranking score. For example, commenting time is standardized and used to calculate the final ranking score.

  If a particular comment has a high score, this indicates that the comment contains more useful information to help the user's decision, such as a shopping decision.

  At reference numeral 150, the comments are ranked based on the score. According to an exemplary embodiment of the present disclosure, comments are displayed in descending order from high score to low score.

  FIG. 3 illustrates an exemplary web page display according to an exemplary embodiment of the present disclosure.

  In the display web page 300, a plurality of display tags 302 are shown at the top of the web page 300 and a list of comments is displayed below the display tag 302. When the user clicks the “all” button 304 on the left side of the display tag 302, all comments are displayed in the sort order according to the present technology. When one of the display tags 302 is clicked by the user, comments corresponding to the corresponding display tags are displayed in the sort order according to the present technology.

  For example, in FIG. 3, when the user clicks the display tag “good quality” 306, the comments corresponding to the display tag “good quality” 306 are displayed in the order according to the method of the present disclosure. For example, the corresponding comment typically includes a combination of a useful attribute 310 such as “product” and an emotional term 320 such as “very good”.

  It should be noted that the specific terms of the useful attributes 310 included in the comments need not be the same. For example, with respect to useful attributes of product quality, one user may refer to “product”, another user may refer to “thing”, and other users may not directly refer to product.

  Further, even if the user expresses the same or similar emotional attitude, the emotional terms or explanations used will not necessarily be the same. For example, one user may use the term “very good”, another user may use the term “very good”, another user may use the term “move as expected”, and so on.

  Therefore, even if the meaning of the combination of the useful attribute 310 and the emotion term 320 found in the comment is the same or similar, the expression will not necessarily be the same. As shown in FIG. 3, regarding useful attributes such as product quality, the comment of a specific user can be “very good product”, “very good product”, “very good product”, “moved as expected”, etc. . According to exemplary embodiments of the present disclosure, such combinations of useful attributes and emotional terms with different specific expressions but substantially the same or similar meaning are displayed to the user after being standardized as shown in FIG. The

  Such an object can be achieved by the display tag 302. The display tag 302 can include two parts. Some are standard useful attributes, and some are standard emotion terms. For example, as shown in FIG. 3, the display tag “good quality” 306 includes two parts. That is, “quality” and “good”. “Quality” is a standard useful attribute, and “good” is a standard emotion term. Therefore, the display tag 302 is a result displayed to the user acquired by standardizing a combination of useful attributes and emotion terms.

  According to an exemplary embodiment of the present disclosure, the user can click on one of the display tags 302 on the web page to view the comment corresponding to a particular display tag. Comments are displayed in the order of sorting or ranking according to the method of the present disclosure.

  According to another exemplary embodiment of the present disclosure, the portion corresponding to the display tag can be highlighted in the comment list.

  In this way, the present technology not only ranks the comments in consideration of the useful attributes of the object, but also displays the ranked results according to different display tags so that the user can make a decision regarding shopping quickly. Do further help.

  Note that the display tag as shown in FIG. 3 is not limiting and is merely exemplary. Based on the description of the present disclosure, those skilled in the art will design different display tags depending on the particular application.

  FIG. 4 is a diagram illustrating an exemplary apparatus 400, such as a computing device, that performs comment ranking by a search engine, according to an exemplary embodiment of the present disclosure.

  The apparatus 400 can include one or more processors or data processing units 402 and a memory 404. The memory 404 stores therein a plurality of modules including an extraction module 410, an acquisition module 420, a determination module 430, a scoring module 440, and a ranking module 450. The extraction module 410 extracts one or more terms that describe the object from comments about the object. The acquisition module 420 acquires one or more useful attributes included in the comment according to a term describing the object. The determination module 430 determines one or more scoring indicators for scoring comments according to the number of useful attributes. The scoring module 440 scores the comments according to the scoring index. The ranking module 450 ranks the comments according to the score.

  According to an exemplary embodiment of the present disclosure, the determination module 430 further determines a scoring index for scoring the comment that includes the number of useful attributes in the comment if the number of useful attributes in the comment is one or more. A first decision submodule may be included.

  According to an exemplary embodiment of the present disclosure, the scoring index for scoring a comment may further include at least one of a comment readability, a rank attribute factor of useless attributes in the comment, and a comment grant time. .

  According to an exemplary embodiment of the present disclosure, the determination module 430 further includes one or more scoring indices other than the useful attributes in the comment if the number of useful attributes in the comment is zero. A second determination sub-module that determines

  According to the exemplary embodiment of the present disclosure, the scoring index other than the index of the useful attribute may include at least one item among the readability of the comment, the rank factor of the useless attribute in the comment, and the comment granting time. .

  According to an exemplary embodiment of the present disclosure, the apparatus 400 may further comprise a display module that displays one or more display tags to the user. The display tag can include standard useful attributes and standard terms describing the object. When the display tag is clicked by the user, the list of user comments corresponding to the display tag is displayed in a rearranged state.

  According to an exemplary embodiment of the present disclosure, the object can be a product, service, or merchant.

  The technology is generally applicable to e-commerce activities. However, the present technology is not limited to such use, and can be implemented on other Internet uses such as news websites, social networks, etc., as long as it is an arbitrary application including comments on a specific matter or phenomenon, that is, an object. Is possible. Those skilled in the art will immediately come up with other Internet applications of the present disclosure.

  It should be understood by one of ordinary skill in the art that the embodiments of the present disclosure may be provided as a product of a method, system, or computer program. Thus, the present technology can be implemented in the form of hardware, software, or a combination of hardware and software. Further, the present disclosure may be implemented as a product of a computer program stored in a computer readable storage medium including, but not limited to, disk memory, CD-ROM or optical memory with computer-executable instructions.

  The foregoing embodiments are merely exemplary embodiments of the present disclosure and are not intended to limit the scope of the present disclosure. The present disclosure will be subject to various modifications and changes by those skilled in the art. Any modification, replacement, or improvement falling within the spirit and principle of this disclosure is intended to be within the scope of this disclosure.

Claims (20)

Extracting one or more terms describing the object from comments about the object;
Determining one or more scoring indicators for scoring the comment according to the number of useful attributes in the comment based on the one or more terms;
Scoring the comment according to the one or more scoring indicators;
Ranking the comments according to the scoring results. Extracting the one or more terms describing the object from the comment about the object;
Perform word segmentation of the comment,
Searching for one or more emotion terms from the result of the word splitting;
Each emotion term is a term that reflects a user's view regarding each useful attribute of the object, and each useful attribute is information that describes an aspect or characteristic of the object.
The method of claim 1.   The method of claim 2, further comprising locating the respective useful attributes corresponding to the respective emotion terms in response to the respective emotion terms.   The method of claim 2, further comprising searching the comment to locate the respective useful attribute corresponding to the respective emotional term.   The method of claim 2, further comprising ignoring the respective emotional term upon receiving a determination that the respective useful attribute corresponding to the respective emotional term has not been found. Standardize one or more useful attributes,
Standardize one or more emotional terms,
The one or more useful attributes;
The one or more standard useful attributes;
The one or more emotional terms;
The method of claim 2, further comprising maintaining at least two correspondences among the one or more standard emotion terms.   The method of claim 2, wherein the finding the one or more emotion terms from the result of the word splitting comprises finding the one or more emotion terms from a list of emotion terms maintained on a server.   The method of claim 1, further comprising data mining one or more other comments about the object to obtain the useful attributes.   The method of claim 1, further comprising presetting the useful attributes for the object.   The method of claim 1, wherein the useful attributes are different for different objects. Depending on the number of the useful attributes in the comment based on the term, the determining one or more scoring indicators for scoring the comment is:
Determining the one or more scoring indicators that include the number of the useful attributes in the comment if the number of the useful attributes is one or more;
The method of claim 1. The one or more scoring indicators are further
Readability of the comment,
A rank factor of one or more useless attributes in the comment;
The method according to claim 11, comprising at least one of commenting times. Depending on the number of the useful attributes in the comment based on the term, the determining one or more scoring indicators for scoring the comment is:
Determining the one or more scoring indicators including one or more indicators other than the useful attributes when the number of the useful attributes is zero;
The method of claim 1. The one or more indicators other than the useful attribute are:
Readability of the comment,
A rank factor of one or more useless attributes in the comment;
The method according to claim 13, comprising at least one of commenting times.   The method of claim 1, further comprising displaying one or more display tags, wherein the at least one display tag includes standard useful attributes and standard terms describing the object.   16. The method of claim 15, further comprising displaying a ranked list of comments corresponding to each display tag selected by the user. An extraction module that extracts one or more terms describing the object from comments about the object;
An acquisition module for obtaining a number of useful attributes in the comment based on the one or more terms;
A determination module that determines one or more scoring indicators for scoring the comment according to the number of the useful attributes in the comment;
A scoring module for scoring the comment according to the one or more scoring indicators;
A ranking module for ranking the comments according to the scoring results.   The apparatus of claim 17, further comprising a display module that displays one or more display tags to a user, each display tag including standard useful attributes and standard terms describing the object. A first determining sub-module that determines the one or more scoring indicators including the number of the useful attributes in the comment if the number of the useful attributes is one or more;
A second determining sub-module that determines the one or more scoring indices including one or more indices other than the useful attributes when the number of the useful attributes is 0;
The one or more indicators other than the useful attribute are:
Readability of the comment,
A rank factor of one or more useless attributes in the comment;
Including at least one of the comment grant times,
The apparatus of claim 17. One or more computer storage media storing computer-executable instructions executable by one or more processors, wherein an operation is performed;
Extracting one or more terms describing the object from comments about the object;
Determining one or more scoring indicators for scoring the comment according to the number of useful attributes in the comment based on the one or more terms;
Scoring the comment according to the one or more scoring indicators;
The one or more computer storage media that perform operations including ranking the comments according to the scoring results.